MARKET EVALUATION INVESTIGATION DEVICE AND INVESTIGATION METHOD
By adopting a component index with which it is possible for the evaluation subject to freely move over a screen and cite components of interest of an evaluation object overall image, it is possible to efficiently acquire information of the components which receive attention through the eyes of a typical consumer without being constrained by components which the specialists have prepared as objects to be evaluated, and by organizing and grouping the information thus acquired and displaying together with positions and frequencies of components of interest, it is possible to easily comprehend investigation results.
Latest CREATIVE BRAINS, K.K. Patents:
An object of the present invention is to provide a new investigation device for market investigation or the like unavailable in the past, capable of efficiently gathering and analyzing or the like objective evaluations of typical consumers or the like as evaluation subjects for qualitative information regarding an object for evaluation such as a product design, which fundamentally cannot be uniquely and quantitatively expressed with numerical values.
BACKGROUND ARTIn recent years, along with higher level performance of various products, generalization of technology, and the globalization, acceleration and the like of information transfer, the importance of “design” which affects product success and failure has increased. However, at the definition phase and modification phase and the like of a “design,” it is difficult to evaluate a “design.” To begin with, the “design” which is the external appearance of a product should be evaluated by the product user (in most cases, the general consumer), and the fact that there are many cases when the evaluation of the “design” based on the sensibility of the general consumer differs from the evaluation by an expert in design or the like is one major reason that “design” evaluation is difficult.
However, as the conventional “design evaluation method,” quantitative investigations such as survey investigations are often used. For these quantitative investigations, not just simple methods of simply marking a circle or an X on choices prepared in advance, but also methods using reply results according to the language of the general consumer have been developed. In specific terms, an item has been proposed whereby an expert prepares in advance a large amount of evaluation language for evaluating a design such as “modem,” “flashy,” or the like, these evaluation language and product photographs are simultaneously presented to a general consumer, for example, and the results replied in language by the general consumer regarding how “modern” or “flashy” that product is are consolidated using a statistical calculation process, to calculate the design evaluation overall points.
However, though this kind of improved quantitative investigation has the advantage of being able to efficiently fetch investigation results from a large number of evaluation subjects using a computer network such as the internet or the like as noted in Patent Documents 1 and 2, it was difficult to obtain investigation results based on the sensibility of general consumers. Specifically, a “design” is created with the subjective intention of a designer who is an expert, but its evaluation and the decision of whether or not to purchase it is made according to the sensibility of a general consumer, and the fact is that there is quite a large skew in the evaluation between experts and general consumers. Because of that, with “design” market investigations, it is that much more important to have the evaluation of a “design” be done not by an expert but rather according to the pure sensibilities of general consumers. However, even when trying to obtain an evaluation according to the pure sensibility of a general consumer, as described previously, with a quantitative study, by the expert preparing in advance many measures describing the “design” such as “modern,” “flashy” or the like, the framework of the design evaluation by the general consumer is restricted within the frame thought of by the expert. Therefore, when there is an evaluation measure in the sensibility of the general consumer that was not thought of by the expert, it is very difficult to detect the evaluation results by that evaluation measure using that quantitative investigation.
Separate from this kind of quantitative investigation, qualitative investigations such as group interviews, depth interviews and the like are also known. Qualitative investigations are a method that in contrast to quantitative investigations that use question sheets or the like, do individual or group interviews with general consumers, and without preparing evaluation language in advance, it is possible to obtain replies which are a “design” evaluation with free language of the consumer. However, with this kind of qualitative investigation, in addition to it being necessary to have an interviewer who has received specialized training have a face to face investigation with the general consumer, there is also the big problem of, realistically, the interviewer being limited to a small number of respondents to the degree he is able to organize them in his head. Also, since the number of respondents is limited to a level of several tens or at the most several hundred people, the reproducibility of the results when reinvestigating is low, and in particular, with products for which the number of general consumers is large with a subject such as cell phones, automobiles, home electronic products and the like, having the objective reliability of the investigation results be even one level lower is unavoidable. In fact, even with interviewers who have received training so as not to use leading questions, since it is impossible to completely prevent the mixing in of the subjectivity of the interviewer, it is difficult to say that the obtained opinion of the respondent is based purely on the sensibility of the respondent, it is not possible to eliminate the risk of the opinion having been drawn out by the interviewer, and in light of that point as well, there are questions about the objective reliability of the investigation results.
BACKGROUND ART DOCUMENTS [Patent Documents]Patent Document 1: JP-A-2002-149900
Patent Document 2: JP-A-2006-244049
SUMMARY OF THE INVENTION Problem the Invention Attempts to SolveThe present invention has been developed with the circumstances as described above as the background, and it is one object: of the present invention to provide the following matters of (i)-(iii) as for evaluation objects for which measurement with numerical values using a precise scale is fundamentally unsuitable, such as for clothing design, cell phone design or the like:
(i) to make it possible to fetch results for which a qualitative evaluation was performed based on the free will of the evaluation subject, without advance preparation of measures or terms for evaluating the object, while removing to the extent possible the mixing in of the subjectivity of an expert,
(ii) to realize high reliability and generalization of evaluation results by efficiently fetching evaluation results by a large number of evaluation subjects without requiring, a large amount of effort or time, and
(iii) to realize a new “market evaluation investigation device and investigation method” that allows display of the obtained evaluation results in a way that is easy to understand.
Means for Solving the ProblemA first mode of the present invention relates to the “market evaluation investigation device” created to attain the object, and provides a market evaluation investigation device being characterized by comprising: (a) an evaluation object display unit for displaying an image of an evaluation object on a screen; (b) component index display unit for displaying a component index specifying in partial form the evaluation object on the screen on which the entire image of the evaluation object is displayed; (c) a component of interest input unit for having the component index moved on the screen and having position information of component of interest input: (d) an information storage unit for storing component of interest-position-information input by the component of interest input unit, respectively fetched by a plurality of evaluation subjects regarding the evaluation object; (e) a grouping calculation unit for grouping together items for which the positions of the components of interest are close, with the component of interest-position-information stored in the information storage unit as the subject; (f) a component of interest position display unit for displaying on the screen the positions of the components of interest for which the number was made smaller by the component of interest-position-information being grouped by the grouping calculation unit; and (g) a component of interest frequency display unit that differentiates a display format of the position of the component of interest according to the number of component of interest-position-information gathered in each component of interest. group, regarding the position of each component of interest displayed on the screen by the component of interest position display unit.
With the investigation device constituted according to this kind of mode, by using the constitutions (a) to (c) noted above, it is possible for the evaluation subject to freely selectively specify components of interest using his own sensibilities. Therefore, it is possible to freely specify the component one is focusing on when the evaluation subject such as a general consumer or the like is doing an evaluation without being hound to evaluation object components prepared in advance by an expert, and it is possible to obtain objective evaluation results for the components being focused on in terms of design.
In the investigation device of this mode, in conjunction with making it possible to freely specify the focus components that the evaluation subject selected himself using the constitutions (a) to (c) noted above, the constitutions (e) to (f) noted above are also employed. This makes it possible to efficiently arrange each separate component of interest obtained by a plurality of evaluation subjects, and to represent the evaluation results in an easy to understand format. In particular, almost none of the position information of each component of interest for which it was made possible to specify freely by each evaluation subject using constitutions (a) to (c) have the same position, but it is possible to efficiently arrange those diverse position information using constitution (e), and to do an easy to understand screen display using constitution (f), and it is also possible to effectively reflect the sensibilities of each evaluation subject arranged using constitution (g) in the form of a frequency and display it on a screen in a way that is easy to understand.
The investigation device of this mode makes also it possible for the evaluation subject himself to perform the evaluation input operation using the constitutions (a) to (c) noted above, as well as for processing the evaluation information calculation process and the evaluation results output process by using the constitutions (d) to (g) noted above. Therefore, the manpower and labor required for the investigation is reduced, and it is possible to efficiently obtain and process evaluation results by a large number of evaluation subjects.
A second mode of the present invention relates to the “market evaluation investigation device” according to the first mode wherein mode, further comprising: a subject attribute input unit by which subject attribute information relating to attributes of the evaluation subject is made to be input is provided; a first evaluation subject selection unit that stores the subject attribute information input using the subject attribute input unit in the information storage unit, and selects the evaluation subject to be the subject using the subject attribute information stored in the information storage unit; a first evaluation information limiting unit thin selects the component of interest-position-information obtained by the evaluation subject selected using the first evaluation subject selection unit from among the component of interest-position-information stored in the information storage unit respectively fetched by the plurality of the evaluation subjects regarding the same evaluation object, wherein processing is executed using the grouping calculation unit with the component of interest-position-information selected using the first evaluation information limiting unit as the subject.
With the investigation device of this mode, as attributes of the evaluation subjects, it is possible to easily obtain investigation results using subject attribute information such as gender, age level, residence, job category, academic background, annual income, amount expended for a specific object, hobbies and the like, for example. Specifically, since it is possible to analyze market evaluation investigation results by narrowing down the evaluation subjects using specific attributes using the first evaluation information limiting unit, and to display those results on a screen, it is possible to easily and efficiently grasp the evaluation differences and the like by the differences in evaluation subject attributes.
The third mode of the present invention relating to the “market evaluation investigation device” according to the first or second mode, further comprising: an overall evaluation input unit for inputting a selection made from a plurality of choices of overall evaluation information relating to the evaluation object prepared in advance, the overall evaluation information inputted, by the overall evaluation input unit being stored in the information storage unit; a second evaluation subject selection unit for selecting evaluation subjects using the overall evaluation information stored in the information storage unit; and a second evaluation information limiting unit for selecting the component of interest-position-information obtained by the evaluation subject selected using the second evaluation subject selection unit from among the component of interest-position-information stored in the information storage unit fetched respectively by the plurality of evaluation subjects regarding the same evaluation object, wherein processing by means of the grouping calculation unit is executed with the component of interest-position-information selected by the second evaluation information limiting unit as the subject.
With the investigation device of this mode, as the overall evaluation regarding the evaluation object, for example, it is possible to distinguish evaluation subjects self-judged as “Like Overall” and evaluation subjects self judged as “Dislike Overall,” analyze the market evaluation investigation results each separately, and display those results on a screen. Because of that, even with the same component of interest, it is possible to do an evaluation distinguishing between whether that is a component of interest that was affected by one's own judgment of “Like Overall” or a component of interest that was affected by one's own judgment of “Dislike Overall.”
The fourth mode of the present invention relating to the “market evaluation investigation device” according to any one of the first through third modes, wherein the grouping calculation unit includes a grouping threshold value setting unit that changes a threshold value which is a maximum separation value of the component of interest-position-information gathered in the same group.
With the investigation device of this mode, it is possible to make more diverse use of analysis results and perform observations more easily by setting the threshold value high and roughly specifying the component of interest with the grouping calculation unit, or setting that threshold value low and observing in detail the differences in the components of interest.
The fifth mode of the present invention relating to the “market evaluation investigation device” according to any one of the first through fourth modes, wherein the grouping calculation unit includes a display group selection unit that selects only groups for which the number of the component of interest-position-information gathered in the same group exceeds a preset standard number, and displays on the screen the component of interest with the component of interest position display unit.
In this mode, the investigation device is able to adjust the standard number of selected component of interest groups to be displayed on the screen. This makes it possible, for example, to make it even easier to understand the results with the analysis results for which components of interest for which only a slight number of evaluation subjects are interested are ignored by raising the standard number, or alternatively to do an even more detailed evaluation by lowering the standard number and making the components of interest of only a small number of evaluation subjects the object.
The sixth mode of the present invention relating to the “market evaluation investigation device” according to any one of the first through fifth modes, further comprising: a language information input unit for inputting a language expressing a reason for the component of interest input with the component of interest input unit; a language display unit for storing language information input using the language information input unit respectively fetched by a plurality of evaluation subjects for the same evaluation object in the information storage unit, and displaying on the screen together with the component of interest the language input using the language information input unit; a display language selection unit for determining the language to display on the screen using the language display unit by selecting only the language input by a number that exceeds a preset standard number among the language input using the language information input unit; and a language frequency information display unit for differentiating the display format of the component of interest according to the number of inputs of each language for each language displayed on the screen with the language display unit.
With the investigation device of this mode, the evaluation results according to the sensibilities of the evaluation subjects can be inputted by using language which is a varied expression means. This makes it possible to obtain extremely detailed evaluation information compared to when simply specifying only “components of interest,” and to perform analysis of a market evaluation that reflects that. Particularly with this mode, the same as with the specifying operation of the “component of interest” described previously, even when doing a “language” input operation, the existence of an expert such as an interviewer or the like is unnecessary, and it is possible for the evaluation subject such as a general consumer or the like to input language freely according to his own sensibility. Owing to this arrangement, it is possible to obtain language information honestly reflecting the sensibility of evaluation subjects such as general consumers or the like as evaluation information, without being bound to evaluation scales or criteria prepared in advance by an expert. In fact, with this mode, by fetching “language” information as evaluation information, qualitative evaluation information more reliably reflecting the qualitative evaluation of the evaluation object is obtained, and it is possible to reflect that in the evaluation results.
The seventh mode of the present invention of the “market evaluation investigation device” according to any one of the first through sixth modes, further comprising: a language information input unit for inputting a language expressing a reason for the component of interest input with the component of interest input unit; and a display matrix calculation unit that, regarding the evaluation object, uses the component of interest-position-information stored in the information storage unit and the language information fetched with the language information input unit, finds a multidimensional matrix that uses as an element each value of a frequency distribution represented having the component of interest-position-information be either of row items or column items, and the language information be the other of the row items or the column items, and obtains a display matrix for which the multidimensional matrix is dimensionally compressed; and a coordinate system association display unit that associates with a coordinate system each component of interest image and each language of the language information of the evaluation object corresponding to the component of interest-position-information based on the display matrix obtained with the display matrix calculation unit, and displays this on the screen.
The investigation device of this mode is able to comprehensively hand component of interest information displayed using images and component of interest information displayed using language and implementing special calculation processing, rather than numerical values or the like, by associating those information using images and information using language together in the same coordinate system and visually developing the display. Therefore, the information using images and the information using language can mutually supplement each other together, and it is possible to display the analysis results of the market investigation in an easier to understand format.
The coordinate system represented CM the screen particularly with this mode does not have a specific measure or the like set prepared in advance, but rather is an item for which image and language information relating to the component of interest is obtained as specific calculation processing results handled comprehensively, and does not have scales or indexes given in advance. Thus, while referencing language associated with an appropriate position using an unknown scale hidden in the depths of these between each component of interest of the evaluation object from images of each component of interest associated with that coordinate system, it is possible to have an objective and visual grasp of each of those components of interest and evaluate them. In other words, the analysis results displayed on the screen with this mode are the results obtained by the ultimate technique for representing qualitative evaluation results based on the sensibilities of evaluation subjects such as general consumers or the like, while removing to the extent possible frameworks and scales given by an expert.
The eighth mode of the present invention relating to the “market evaluation investigation device” according to the seventh mode, wherein in the display matrix calculation unit, the component of interest-position-information and the language information respectively fetched for a plurality of the evaluation objects are consolidated into one multidimensional matrix, and the display matrix is found from the multidimensional matrix.
With the investigation device of this mode, for the plurality of evaluation objects, it is not necessary to directly fetch their reciprocal evaluation information, and based on the evaluation information obtained for each of the evaluation objects (including component of interest-position-information and language information), their reciprocal evaluation results are found using the previously described special calculation process, and it is also possible to display those on the screen in an easy to understand format.
The ninth mode of the present invention relating to the “market evaluation investigation device” according to the seventh or eighth mode, wherein in the display matrix calculation unit, an overall image of the evaluation object is added to the same row or column item as the component of interest-position-information in the multidimensional matrix, and a total value of each frequency of the language information used as an element of the multidimensional matrix is used as an element of the overall image, while the display matrix is obtained by dimensionally compressing the multidimensional matrix obtained using that, and in the coordinate system association display unit, in addition to the component of interest image and the language, the overall image is also associated with the coordinate system, and is displayed on the screen.
With the investigation device of this mode, by doing analysis including the entire image of the evaluation object, it is possible to display on the screen the reciprocal evaluation results of a plurality of evaluation objects in an even easier to understand format. Also, by observing which relative position each component image constituting each overall image is associated with in the coordinate system, it is easy to understand things such as how each component of interest expressed by each component image contributes to the overall evaluation of the evaluation object.
The tenth mode of the present invention relating to the “market evaluation investigation device” according to the ninth mode, wherein the coordinate system association display unit includes a display selection unit of the component of interest image that selectively hides the display of the component of interest image on the screen.
With the investigation device of this mode, for example when associating evaluation results for a plurality of evaluation objects to one coordinate system, it is possible to resolve problems such as having there be more images than necessary and having it be difficult to understand the subject evaluation results.
The eleventh mode of the present invention relating to the “market evaluation investigation device” according, to any one of the seventh to tenth modes, further comprising an element correlation display unit which, by correlating each item of the component of interest-position-information in the multidimensional matrix with each language of the language information used as an element of each of the items and displaying this on the screen, displays the correlation of the component of interest image and each language displayed on the screen with the coordinate system association display unit.
With the investigation device of this mode, for example even when images and language are associated for a plurality of components of interest with one coordinate system, it is possible to specify language correlated to a specific component of interest and make it easier to understand.
Another aspect of the present invention relating to the “market evaluation investigation method” provides a market evaluation investigation method being characterized by comprising: (α) an evaluation object display step of a computer displaying an image of an evaluation object on a screen; (β) a component of interest-position-information fetching step of the computer displaying a component index that indicates in parts the evaluation object on the screen on which the entire image of the evaluation object is displayed, and fetching position information of component of interest by fetching the position information on the screen of the component index moved on the screen by outside input; (χ) a storage step of the computer storing in an information storage unit the component of interest-position-information fetched with the component of interest-position-information fetching step; (δ) a grouping calculation step of the computer using the component of interest-position-information stored in the information storage unit as the subject, and executing a calculation process that gathers and groups items with nearby positions of the components of interest; and (ε) a display step of the computer displaying on the screen the positions of the components of interest made into a small number by being grouped with the grouping calculation step, and displaying a frequency belonging to each of the components of interest by differentiating the display format of the position of the component of interest according to the number of the component of interest-position-information gathered in each component of interest group.
According to the investigation method of the present invention, the same as with the investigation device of the first mode, the components of interest in terms of design selected based on the personal sensibilities of the evaluation subject such as a general consumer or the like is reflected, and it is possible to efficiently obtain objective evaluation results with little effort.
With the investigation method of the present invention, it is preferable to further use (φ) a language information fetching step of the computer fetching from outside language information constituted from language representing a reason for interest regarding the component of interest, (γ) a storage step of the computer affiliating the language information fetched with the language information fetching step to the component of interest-position-information fetched with the component of interest-position-information fetching step and storing this in the information storage unit, (η) a display matrix calculation step of the computer using the component of interest-position-information and the language information stored in the information storage unit, finding a multidimensional matrix using as an element each value of a frequency distribution represented. having the component of interest-position-information be either of row items or column items and having the language information be the other of the row items or the column items, and finding a display matrix for which the multidimensional matrix is dimensionally compressed, and (ι) a coordinate system association display step of the computer, based on the display matrix obtained, with the display matrix calculation step, associating each component of interest image of the evaluation object corresponding to the component of interest-position-information and each language of the language information with a coordinate system, and displaying that on the screen.
In this way, by using each of the aforementioned steps (α) through (ι), the same as with the investigation device according to the seventh mode, with the coordinate system that is the coordinate system obtained by considering and analyzing the mutual relationship between the information by images and the information by language, which is expressed using an unknown scale hidden in the depths between each component of interest of the evaluation object, it is possible to display on the screen as market evaluation analysis results the screen information for which their information by image and information by language are suitably associated. Accordingly, frameworks or scales by an expert are excluded to the extent possible, and the qualitative evaluation results based on the sensibilities of the evaluation subjects such as general consumers or the like can be displayed in an easy to understand format.
Yet another aspect of the present invention relates to an information recording medium which can be read by a computer on which a program is placed For executing on the computer the “market evaluation investigation method” of the present invention as described above, or alternatively a program for executing on a computer that “market evaluation investigation method.”
EFFECT OF THE INVENTIONAs is clear from the description above, with the market evaluation investigation device and investigation method of the present invention, it is possible to efficiently fetch results for which a qualitative evaluation was performed based on the free will of the evaluations subject while excluding to the extent possible the mixing in of the subjectivity of an expert, and to display these in an easy to understand way while retaining high reliability and generalization of the evaluation results.
Following, we will describe embodiments of the present invention while referring to the drawings.
First,
Also,
(i) An evaluation object image information input unit 26 for inputting image data of the evaluation object.
(ii) A component of interest input unit 28 as the component of interest input unit for inputting the position information of the evaluation object component of interest.
(iii) A subject attribute information input unit 30 as the subject attribute input unit for inputting attribute information of the evaluation subject.
(iv) A first evaluation subject selection input unit 32 as the first evaluation subject selection unit for selecting evaluation subjects and inputting subject attribute information to be subject in order to narrow down the analysis subject information.
(v) An overall evaluation input unit 34 as the overall evaluation input unit for inputting the overall evaluation of the evaluation object.
(vi) A second evaluation subject selection input unit 36 as the second evaluation subject selection unit for selecting the evaluation subjects and inputting the overall evaluation information to be subject in order to narrow down the analysis subject information.
(vii) A threshold value input unit 38 as the grouping threshold value setting unit for inputting the threshold value (tp) when gathering and grouping position information of the component of interest of the evaluation object.
(viii) A component of interest standard number input unit 40 for inputting the standard number (tn) in order to narrow down the components of interest displayed when displaying on the screen the position of grouped components of interest.
(ix) A grouping calculation unit 42 as the grouping calculation unit for gathering and making into a small number the components of interest with nearby positions, while referring to the threshold values (tp) input with the threshold value input unit 38.
(x) A first evaluation information limiting unit 44 as the first evaluation information limiting unit for doing limiting processing of analysis subject information based on the subject attribute information input with the first evaluation subject selection input unit 32.
(xi) A second evaluation information limiting unit 46 as the second evaluation information limiting unit for doing limiting processing of analysis subject information based on the overall evaluation information input with the second evaluation subject selection input unit 36.
(xii) A display group selection unit 48 as the display group selection unit for selecting the item to be displayed from among the components of interest grouped with the grouping calculation unit 42, while referring to the standard number (tn) input with the component of interest standard number input unit 40.
(xiii) A component of interest-position-information storage unit 50 for storing the position of the component of interest input with the component of interest input unit 28.
(xiv) A subject attribute information storage unit 52 for storing subject attribute information input with the first evaluation subject selection input unit 32.
(xv) An overall evaluation information storage unit 54 for storing overall evaluation information input with the overall evaluation input unit 34.
(xvi) A calculation results storage unit 56 for storing calculation results information found with the grouping calculation unit 42 or the like.
(xvii) An evaluation object display unit 58 as the evaluation object display unit for displaying on the screen the evaluation object input with the evaluation object image information input unit 26.
(xviii) A component index display unit 60 as the component index display unit for displaying on the screen a component index for specifying the component of interest on the overall image of the evaluation object displayed with the evaluation object display unit 58.
(xix) A component of interest position display unit 62 as the component of interest position display unit for displaying on the screen the component of interest selected with the display group selection unit 48.
(xx) A component of interest frequency display unit 64 as the component of interest frequency display unit for displaying the number (level) of the components of interest gathered with the grouping calculation unit 42 for each component of interest displayed on the screen with the component of interest position display unit 62.
Following, using the constitutions as shown in
First, when start is instructed and the analysis process starts, the computer implements the information fetching process shown in
With this embodiment, we showed an example using a women's clothing design as the evaluation object, but as long as the evaluation object of the present invention is an item requiring qualitative evaluation, anything is acceptable as long as it can be identified using images. For example, in addition to clothing, of course with items such as automobiles, televisions, sweets, cuisine and the like, and also with advertisements, image films and the like, as long as it is possible to identify that evaluation object using images, and it is possible to select components of interest of a qualitative evaluation with that image, it is possible to be an evaluation object of the investigation device of the present invention.
Also, the image information used with the present invention is acceptable as long as the contents are image data, that it is possible to understand the design of the evaluation object, and also that it is an item for which it is possible to reciprocally distinguish and identify their design when there is a plurality of evaluation objects, and for example in addition to photographic data, it is also possible to use an item created using graphic software or the like. Also, the image data can also be animated data, but still image data is preferable because it is easy to handle and can be understood quickly.
Then, once the evaluation object image information is fetched, at the subsequent steps S3 to S10, evaluation information by the evaluation subject is fetched regarding. that evaluation object.
To do that, first, at step S3, an image of that. evaluation object is displayed on the screen by the evaluation object display unit 58, and at step S4, the evaluation is fetched for that overall evaluation object. In specific terms, for example as shown in
At the next step S5, the component of interest-position-information for the evaluation object is fetched using the component of interest input unit 28 and the component index display unit 60. In specific terms, as shown in
Also, at step S6, the evaluation subject attribute information is fetched using the subject attribute information, input unit 30. In specific terms, for example as shown in
Then, by executing the judgment and loop processes at steps S7 and S8, for the same evaluation object, the overall evaluation information, the component of interest-position-information, and the subject attribute information according to the aforementioned steps S4 to S6 are fetched, and this is implemented for all the evaluation subjects.
Furthermore, when there is a plurality of evaluation objects, by executing the judgment and loop processes of steps S9 and S10, for all the evaluation objects, the overall evaluation information, the component of interest-position-information, and the subject attribute information according to the aforementioned steps S3 to S8 is fetched for each of the evaluation subjects. With this embodiment, we will describe a case when the number of evaluation objects is one, and the process ends without performing the loop process using steps S9 to S10.
Also, the overall evaluation information, the component of interest-position-information, and the subject attribute information fetched by executing the information fetching process shown in
Subsequently, using the information fetched by the information fetching processes described above, the calculation analysis process shown in
Specifically, with the subject attribute selection value fetched at step T3, for example it is possible to do analysis using information obtained with investigation results for which only women are the subject, do analysis using information obtained with investigation results for which only a specific age level is the subject, and furthermore, to obtain the difference between analysis results when women are the subject and when men are the subject, and the like. Also, with the overall evaluation selection value fetched at step T4, it is possible to do analysis using information obtained with evaluation subjects who sensed “Like” as the overall evaluation for the evaluation object as the subject, to do analysis using information obtained with evaluation subjects who sensed “Dislike” as the subject, or to do analysis using information obtained with evaluation subjects who sensed “Like or Somewhat Like” as the subject.
After narrowing the information as this kind of analysis subject, at step T5, using the component of interest-position-information stored in the component of interest-position-information storage unit 50, grouping calculation is executed for the position of interest information obtained by the evaluation subjects selected as the subject of analysis.
This grouping calculation is processed according to the flow chart shown in
With this joining, a new component of interest position is set as the coordinate value for the center of balance of the two component of interest positions, and also, a number (frequency) of the components of interest is set as an item for which two components of interest exist at that position. Then, by executing steps U1 to U4 with all the position of interest information including the new component of interest set by joining as the subject, the grouping calculation process is repeated until there are no more components of interest positioned nearby at the threshold value (tp) or less. By doing this, the components of interest are gathered, and a suitable number of new components of interest which are grouped as groups of suitable size reflecting the size of the threshold value (tp) is set. As the grouping, calculation process advances, the frequency of the two components of interest joined to each other increases, but coordinate values are determined for the component of interest after joining with that frequency used for weighting. In specific terms, when obtaining a new component of interest C by joining the component of interest A with the frequency (joint count) 10 and the component of interest B with the frequency (joint count) 2, the point for which the distance from A on a straight line connecting the A coordinates and the B coordinates is ⅕ (the distance from B is ⅘) becomes the position (coordinates) of the new component of interest C. and the frequency of that component of interest C is 12.
Working in this way, after the grouping calculation process is completed, at the subsequent step T6 in
Next, at step T7, using the component of interest standard number input unit 40, the component of interest display standard number (tn) is fetched that determines the degree of narrowing down of the components of interest for which the calculation results are displayed on the screen. Then, at the subsequent step T8, while referring to this component of interest display standard number (tn) the display format is determined from the frequency of each component of interest found with the grouping calculation process described above, the component of interest that is to be the display subject and its display format are determined, and at step T9, this is displayed on the screen as the result of the calculation analysis process (specifically, the market evaluation investigation results).
Specifically, with the grouping calculation process at step T5 described above, items for which the position is almost the same for a plurality of components of interest are gathered and regarded as the same, but even if the position is separated, for example with a frequency of 1 (component of interest input by only one person), these remain without being joined. However, when doing a market evaluation investigation, there are runny cases when there is no value in having items for which the frequency is extremely low such as that be a subject of evaluation, and in fact there is the risk of the judgment being distorted for the investigation results if that is displayed on the screen as analysis results. In light of that, at step T8, for components of interest of a frequency smaller than the preset standard number (tn), it is possible to not display them on the screen as investigation results and to ignore them.
Also, even with components of interest having a frequency of the standard number (tn) or greater, there is a difference in terms of the market evaluation importance based on the size of that frequency. Because of that, at step T8, when displaying a plurality of components of interest on the screen, that display is processed so as to be performed in a format reflecting the frequency of each component of interest. In specific terms, for example, when displaying on the screen each component of interest by having a specified area on the overall image of the evaluation object enclosed by frame such as a circle, square or the like, it is possible to have the concentration or brightness, color saturation, sharpness, transparency or the like within that frame be differentiated in gradations according to the frequency of each component of interest. Incidentally, with this embodiment shown in
In this way, according to the market investigation analysis device and analysis method as the first embodiment, it is possible for the general consumer who is the evaluation subject to freely point out the parts he notices (components of interest) only by his own senses. Because of that, in contrast to quantitative investigations such as a conventional survey or the like for which a selection range of answers is prepared in advance, it is possible to obtain investigation results that directly reflect the evaluation results by the sensibility of the replying subject. In fact, the reply of the evaluation subject, by using an internet line or the like, can be obtained by utilizing a very large number of computers, mobile terminals or the like. Because of that, in contrast to quantitative investigations such as a conventional depth interview or the like for which the reply collection volume is limited, it is possible to obtain investigation results that easily and efficiently reflect the replies by evaluation subjects exceeding several thousand or several tens of thousands. Also, the evaluation results processed by the analysis calculation can be simultaneously displayed with the component of interest position and frequency (number) on the screen, and are excellent for evaluation results recognition and use.
Also, with this embodiment, since when obtaining replies, the subject attribute information and overall evaluation information are fetched together, the evaluation subjects are selected using these information, and it is possible to obtain evaluation results with selected specific evaluation subjects as the target, to do a mutual comparison of the evaluation results by different evaluation subjects and the like.
In specific terms, the evaluation results screen shown by example in
Incidentally, it is possible to have input of the selection value of the subject attribute information at step T3 performed by the selection operation using a button 78 on the screen, and also to display on the same screen evaluation results by the calculation process of step T5 reflecting those selection results, and a specific example of this is shown by example in
Furthermore, with this embodiment, when executing the process of displaying on the screen the investigation results with the analysis calculation using the fetched market evaluation information, two threshold values are used. One of those is “the threshold value (tp) fetched at step T2 for controlling the positions seen as equivalent,” and the other is “the standard number tuft which is the threshold value fetched at step T7 to control the number of respondents.” The former threshold value (tp) is a threshold value for judging at what level of error range the center coordinates of the component of interest selected freely by a large number of evaluation subjects is regarded as having selected “the same site,” and the latter standard number (tn) is a threshold value for controlling at about what ratio of people within the number of evaluation subjects a selected site is displayed on the screen as the evaluation result.
Incidentally, it is possible to have inputting of the threshold value (tp) with step T2 performed on a dialog box 80 on the screen, and also to have the evaluation results by the calculation process at step T5 reflecting that input value displayed on the same screen, and a specific example of that is shown by example in
When the results display by the threshold value setting of “10 pixels” shown in
Also, it is possible to have inputting of the standard number (tn) at step T7 be performed on a dialog box 82 on the screen, and also to have the investigation results displayed on the same screen by reflecting that standard number when doing the screen display process at step T8, and a specific example of that is shown by example in
When the results display according to the “3%” standard number setting shown in
With
Next, we will describe the market evaluation investigation device and investigation method as the second embodiment of the present invention while referring to
Specifically, a is clear from the schematic drawing of the hardware configuration shown in
Then, as shown in the flow charts in
In more detail, first, when doing the information fetching process shown in
In specific terms, as shown for example in
Also, the language information input at this step S5-1 is stored in the language information storage unit 92 of the information storage unit 22, and also, when doing the calculation analysis process shown in
In specific terms, as shown for example in
The narrowing of the display language according to the standard number of that language display can be processed by setting in advance the minimum language count to he displayed on the screen or the like, and also for example the same as with the standard number (tn) of the component of interest display at step T7 shown by example with the first embodiment, by inputting using numbers a ratio (%) for which that same language was input among all the evaluation subjects, inputting using a slider operation or the like.
Also, it is possible to display the level of input ratio of that language for all the evaluation subjects so as to be able to easily recognize it on the screen. In specific terms, as shown for example in
Yet further, with this embodiment, using the element correlation display unit 98 (
The specific mode for mutually correlating and displaying each component of interest with each language using the element correlation display unit 98 is not limited to that shown in
Also, with this embodiment as well, being equipped combining the same functions as those of the first embodiment is as described previously, and for example as shown in
Furthermore. with this embodiment, by performing selection and input of only specific components of interest to focus on using the component of interest display selection input, unit 88 (
As described above, according to the market investigation analysis device as the second embodiment of the present invention, by the general consumer who is the evaluation subject freely indicating the part he is interested in (component of interest) using only his own sensibilities, it is possible to avoid to the extent possible the influence of experts and the like, and to efficiently fetch evaluation information reflecting the honest sensibilities of the general consumer, and it is possible to exhibit the same technical effects as those of the first embodiment of being able to display on the screen evaluation results reflecting that in a format that can easily and efficiently be recognized and understood and the like.
In addition to that, with this embodiment, in addition to the same overall evaluation information and component of interest information as that of the first embodiment as the evaluation information of the evaluation object by the evaluation subject, language information is also fetched together. Then, the fetched language information is tallied for each component of interest, and by displaying on the screen the language information with a large input count, this is reflected in the evaluation results, and by doing that, it is possible to more accurately acquire evaluation results by using language to understand the sense of the evaluation subject in relation to the evaluation object.
In particular with this embodiment, a big feature here is that as the language information for evaluating the evaluation object, rather than language prepared in advance to direct someone such as in a conventional survey or the like, language as expressed according to the evaluation subject's own free sensibility is used without providing limits of any kind in a state of simply displaying the overall image of the evaluation object. In this way, with freely expressed language as the subject, finding the commonality of that makes it possible to understand the honest sensibilities of general consumers rather than experts as investigation results.
In fact, with this embodiment, by expressing, sensibilities by the general consumer's free indication of components of interest and free language expression relating to those components of interest, while ensuring qualitative reliability compared with a quantitative investigation, by using as subjects a very large number of general consumers with information gathering and calculation processing using computers, it is possible to ensure quantitative and objective reliability using a volume of information that is comparable to a quantitative investigation.
Furthermore, with this embodiment, for each language associated with a component of interest and displayed on a screen, since it is possible to display on the screen the input count. level (frequency) together, the person viewing the investigation results display screen is able to understand. the differences in the importance level (frequency) between each component of interest together while accurately understanding each component of interest of the evaluation object. Because of that, it is possible to understand the abstract and subjective senses held by the general consumer with language as the medium for each component of interest, and to recognizes by the senses the accuracy and depth of that. understanding with the frequency of each language input for that component of interest as the index.
With this embodiment, in
Also, when inputting language information, there is also concern about the problem of the labor burden for the evaluation subjects, but with this embodiment, that point was also considered, and it has been made easy for each evaluation subject to reply. Specifically, at first, the evaluation object overall image 66 is shown to the evaluation subject, and at that stage, focusing on making it easy to perform the evaluation by one's own senses at the level of “Like/Dislike” for that evaluation object, fetching of the overall evaluation results is performed at the starting stage of step (S4), and after that, particularly for the liked parts (components of interest), the evaluation subject is asked to narrow his focus by his own effort (operation) (S5). After that, again, for those components that were focused on, questions are asked of the reasons and the like, and by having them do text input, the evaluation subject text input can be performed relatively easily Specifically, if replying “Site I Like” or “Site I Dislike” within the overall image, this is relatively easy, because it is easier for words for evaluating those sites to come out naturally alter specifying oneself the “Site I Like” or “Site I Dislike.”
Furthermore, with this embodiment, when fetching language information (step S5-1), it is possible for the evaluations subjects to input language freely with this embodiment, but even when language information is collected rising a method that is a totally free language input, for the language input in a state seeing the image 66 showing the same evaluation object, particularly when the investigation is performed at a scale exceeding several thousand evaluation subjects, the existence of a standard enumeration rate that can be processed quantitatively has already been confirmed by the inventors. In that case, from the purpose of fetching language information that can be processed more efficiently and more precisely, as shown by example in
Also, due to reasons such as there being a low reply count (number of evaluation subjects), or there being a high input character count or the like, there are also cases when the enumeration rate of the same character string (language) is low. In such a case, using thesaurus data, it is possible to do integrated notation of synonyms for the input language, but from the perspective of the precision of the evaluation results faithfully reflecting the sensibilities of the evaluation subjects, this is not necessarily the ideal method. It is more ideal to use a morphological analysis means. The morphological analysis means itself has been known from the past, and for example as software operated on a general use personal computer as a Japanese morphological, analysis system, it is possible to obtain “JUMAN” or “CHASEN” or the like under the copyright of the Nagao Research Laboratory of Kyoto University Engineering Department and the Matsumoto Research Laboratory of Nara Institute of Science and Technology or the like, for example, and it is possible to use those kinds of various types of morphological analysis systems that are available to the public, so here, we will omit a detailed description. Naturally, when fetching the language of this embodiment using a morphological analysis system, it is desirable to eliminate from the subject only parts of speech specifying “particles” or the like. This is because particles and the like do not constitute qualitative information of the evaluation object.
Furthermore, we will describe the market evaluation investigation device and investigation method as the third embodiment of the present invention while referring to
Specifically, as is clear from the schematic diagram of the hardware configuration shown in
Then, with this embodiment, the same as with the second embodiment, after executing the fetching process of each type of information such as the overall image, components of interest, language and the like according to the flow chart in
Also, the information calculation analysis process according to this embodiment is used more optimally for a case when investigating a market evaluation simultaneously for three to several tens of a plurality of evaluation objects more than when investigating a market evaluation having as the subject one or two evaluation objects, so with the description below, as represented in the flow chart in
In more detail, with the information fetching process shown in
Here, for all a types of evaluation Objects (clothing designs), it is also acceptable to do information gathering by having them evaluated by all the evaluation subjects, but there is the problem that when the number (n) of clothing designs is too high, the burden on the evaluation subjects is too high. In light of that, with this embodiment, a list of all the clothing designs is displayed at step S2-1, and at the subsequent step S2-2, the evaluation subject selects several of the items he likes among those (e.g. 5) and inputs the selection information, and only the selected a units of clothing, designs are set as the evaluation objects for that evaluation subject.
With the a units of clothing designs selected in this way as the evaluation objects, at steps S3 to S5-3, for each clothing design as the respective evaluation object, the same as with the second embodiment, the overall evaluation information, component of interest information, and language information is fetched. Then, after the various information is fetched for all the a units of clothing design, the attribute information of that evaluation subject is fetched at step S6, and information fetching by that evaluation subject is completed.
After that, by loop processing with steps S7 and S8, for all of a plurality of evaluation subjects, by executing the aforementioned steps S2-1 to S6, information fetching by each of the evaluation subjects is executed.
For the evaluation information fetched in this way, though it is not impossible for example for a live human analyst to read and interpret this, when the number of evaluation objects becomes large, an extremely large amount of labor and time is required by the analyst, and also, having the analyst's individual subjective view enter the analysis results is unavoidable, In light of that, with this embodiment, according to the flow chart shown in
In specific terms, at steps ST-1 to T7-2 in
Said another way, with the data matrix 114 shown in
Specifically, with the data matrix 114, based on the mutual relative relationship between a plurality of evaluation objects and components of interest, this is a correlation matrix as as multidimensional matrix, representing the relationship of the plurality of evaluation objects and components of interest with a large number of language (morphemes) as the overall item. Then, the display matrix obtained by dimensionally compressing this data matrix is converted to a display matrix on a representation space for which a suitable number of dimensional axes (with this embodiment, two dimensions) used as coordinate axes are expressed using a coordinate system, and with this coordinate system as a representation space, at least a portion of the relationship of the qualitative information of those is displayed in a state replaced as a physical positional relationship with the coordinate system.
Here, this data matrix 114 is a high order item having a great many Characteristic dimensional axes, so to project this in a representation space having a low order of dimensional axes for which the substance of the relationship does not change, preferably a representation space of two dimensions or three dimensions, a method is used that uses the singular value decomposition (S.V.D.) noted above. Specifically, the data matrix 114 uses image data for which evaluation objects and components of interest can be specified as the column items, and also uses all the language fetched for those image data as the row items, and each value of the frequency distribution of the language correlated with each image data is obtained as an element, so this data matrix 114 is generated using a number of columns in the amount of image data corresponding to the number of evaluation objects and components of interest, and the number of rows in the amount of all the language types fetched for each component of interest. With this data matrix, even when the row items and column items are exchanged, there is no effect on the analysis results. Then, when doing dimensional compression of this correlation matrix, first, when the correlation matrix is set with an I row and J column matrix as X, as shown with the formula below, a profile matrix P is found having the element pij for which each element xij of this correlation matrix X is divided by the sum N.
Next, as shown by the formula below, by implementing a similar process to chi-square conversion on each element p of this profile matrix P, a process matrix A having elements a converted based on the separation degree of each element of that profile matrix in relation to the expected values is obtained,
Then, singular value decomposition is implemented on the process matrix A obtained in this way. Specifically, when the matrix obtained using the pre-processing as described previously is used as the I row J column process matrix A, it is possible to do singular value decomposition as noted in the formula below for this process matrix A.
A=UDVT
U is an I row k column matrix, where D is a k row k column diagonal matrix. V is a J row k column matrix, and the superscript T indicates a transposed matrix. Also, with this formula, the k column vectors of the I dimension with the matrix U are (u1, u2, u3, . . . uk), and the k column vectors of the J dimension of the matrix V are (v1, V2, v3, . . . vk), and also, when the matrix D diagonal elements (d11, d22, d23, . . . dkk) are D1≧d2≦d3≧ . . . ≅dk, the following formula is established. The matrix D diagonal elements are d1≧d2≧d3≧ . . . ≧dk>0, and doing this is easy.
However, Qs=usvsT (where s is an integer from 1 to k)
Therefore, from the formula above, matrix A becomes a linear sum of matrix Q, and the characteristic vector corresponding to a large singular value represents the matrix A well. Because of that, by taking in only a suitable number of the singular values from the upper level, it is possible to obtain a low order display matrix U*, V* by converting to as representation space with a low characteristic dimension count from the high order data matrix 114 having a high characteristic dimension count as described previously.
Then, the data of the display matrix obtained in this way is stored in the calculation results storage unit 56 of the information storage unit 22. It is also possible to implement post-processing by adjusting the relative weighting of each dimension axis direction as necessary for the display matrix obtained in this way. In specific terms, it is possible to do adjustment by multiplying a matrix suitably weighted in each dimension axis direction on the display matrix U*, V* obtained with a suitable number of dimensions. It is possible to suitably use a singular value corresponding to each dimension axis for this weighting, for example.
Also, when the display matrix is found with the calculation process with steps V1 and V2, and each element is associated with the coordinate system based on that display matrix, it is possible to display all the elements, but when there are too many display elements, there is the risk of it being difficult to do the display on the screen or to understand it. In that case, at step V3, it is desirable to suitably narrow down the elements (image, component image, language) displayed on the screen. For example, it is also possible to display on the screen only the component of interest images or language having a frequency of a standard number or above based on the standard number fetched at step T7 or T7-2.
In this way, the investigation results displayed on the screen at step T9 are the items shown in
Therefore, by displaying on the screen the display matrix generated using the calculation process described above as the investigation results, a conceptual space for organizing a large number of clothing designs 112 (1 to n) is automatically generated, and it is possible to make this visible in a form for which the relationship between the plurality of clothing design and the correlation between the language as reason items can be intuitively understood. At that time, there is absolutely no need for a person to think of things such as whether the clothing designs can be suitably classified or the like regardless of the coordinate axis used (coordinate axis with what kind of meaning).
Incidentally, when observing the investigation results screen shown in
In this way, according to this embodiment, the “Structural Relationship Between a Plurality of Clothing Designs” that goes beyond the evaluation of individual clothing designs does not require any kind of advance knowledge or reference information, and can be visualized simply from the data in matrix A.
Also, in the display screen of
When observing this
Subsequently,
Compared to the case of “Clothing Design A” 112 (1) shown in
With this screen display shown in
Also, with this embodiment, on the display screen, when a selection operation such as a click or the like is done on the display location of the red shoes [E4], as shown in
By observing this language information, we can see that with a linear sum of reasons such as “Sandal,” “Lovely,” “Color” or the like, the red shoe component of interest diagram (Image) is created (associated). Using this display, “Lovely” has more of a right side a meaning in the concept space displayed on the screen, and “Sandal” can be visualized as being an element having lower side characteristics. Specifically, even with the display screen of the investigation results with the second embodiment (see
Above, we gave a detailed description of the embodiments of the present invention, but the present invention can be implemented in modes with the addition of various modifications, edits, improvements and the like based on the knowledge of persons skilled in the art, and those kinds of various modes are all included in the technical scope of the present invention provided they do not stray from the gist of the present invention.
KEYS TO SYMBOLS26: Object for evaluation image information input unit, 28: Component of interest input unit, 30: Subject attribute information input unit, 32: First evaluation subject selection input unit, 34: Overall evaluation input unit, 36: Second evaluation subject selection input unit, 38: Threshold input unit, 40: Component of interest reference number input unit, 42: Grouping calculation unit, 44: First evaluation information limiting unit, 46: Second evaluation information limiting unit, 48: Display group selection unit, 50: Component of interest-position-information storage unit, 52: Subject attribute information storage unit, 54: Overall evaluation information storage unit, 56: Calculation result storage unit, 58: Object for evaluation display unit, 60: Component index display unit, 62: Component of interest position display unit, 64: Component of interest frequency display unit
Claims
1. A market evaluation investigation device comprising:
- an evaluation object display unit for displaying an image of an evaluation object on a screen;
- a component index display unit for displaying a component index specifying in partial form the evaluation object on the screen on which the entire image of the evaluation object is displayed;
- a component of interest input unit for having the component index moved on the screen and having position information of component of interest input;
- an information storage unit for storing component of interest-position-information input by the component of interest input unit, respectively fetched by a plurality of evaluation subjects regarding the evaluation object;
- a grouping calculation unit for grouping together items for which the positions of the components of interest are close, with the component of interest-position-information stored in the information storage unit as the subject;
- a component of interest position display unit for displaying on the screen the positions of the components of interest for which the number was made smaller by the component of interest-position-information being grouped by the grouping calculation unit;
- a component of interest frequency display unit that differentiates a display format of the position of the component of interest according to the number of component of interest-position-information gathered in each component of interest group, regarding the position of each component of interest displayed on the screen by the component of interest position display unit.
2. The market evaluation investigation device according to claim 1, further comprising:
- a subject attribute input unit by which subject attribute information relating to attributes of the evaluation subject is made to be input is provided;
- a first evaluation subject selection unit that stores the subject attribute information input using the subject attribute input unit in the information storage unit, and selects the evaluation subject to be the subject using the subject attribute information stored in the information storage unit; and
- a first evaluation information limiting unit that selects the component of interest-position-information obtained by the evaluation subject selected using the first evaluation subject selection unit from among the component of interest-position-information stored in the information storage unit respectively fetched by the plurality of the evaluation subjects regarding the same evaluation object;
- wherein processing is executed using the grouping calculation unit with the component of interest-position-information selected using the first evaluation information limiting unit as the subject.
3. The market evaluation investigation device according claim 1, further comprising:
- an overall evaluation input unit for inputting a selection made from a plurality of choices of overall evaluation information relating to the evaluation object prepared in advance, the overall evaluation information inputted by the overall evaluation input unit being stored in the information storage unit;
- a second evaluation subject selection unit for selecting evaluation subjects using the overall evaluation information stored in the information storage unit; and
- a second evaluation information limiting unit for selecting the component of interest-position-information obtained by the evaluation subject selected using the second evaluation subject selection unit from among the component of interest-position-information stored in the information storage unit fetched respectively by the plurality of evaluation subjects regarding the same evaluation object,
- wherein processing by means of the grouping calculation unit is executed with the component of interest-position-information selected by the second evaluation information limiting unit as the subject.
4. The market evaluation investigation device according claim 1, wherein the grouping calculation unit includes a grouping threshold value setting unit that changes a threshold value which is a maximum separation value of the component of interest-position-information gathered in the same group.
5. The market evaluation investigation device according claim 1, wherein the grouping calculation unit includes a display group selection unit that selects only groups for which the number of the component of interest-position-information gathered in the same group exceeds a preset standard number, and displays on the screen the component of interest with the component of interest position display unit.
6. The market evaluation investigation device according claim 1, further comprising: a language information input unit for inputting a language expressing a reason for the component of interest input with the component of interest input unit; a language display unit for storing language information input using the language information input unit respectively fetched by a plurality of evaluation subjects for the same evaluation object in the information storage unit, and displaying on the screen together with the component of interest the language input using the language information input unit; a display language selection unit for determining the language to display on the screen using the language display unit by selecting only the language input by a number that exceeds a preset standard number among the language input using the language information input unit; and a language frequency information display unit for differentiating the display format of the component of interest according to the number of inputs of each language for each language displayed on the screen with the language display unit.
7. The market evaluation investigation device according to claim 1, further comprising: a language information input unit for inputting a language expressing a reason for the component of interest input with the component of interest input unit; and a display matrix calculation unit that, regarding the evaluation object, uses the component of interest-position-information stored in the information storage unit and language information fetched with the language information input unit, finds a multidimensional matrix that uses as an element each value of a frequency distribution represented having the component of interest-position-information be either of row items or column items, and the language information be the other of the row items or the column items, and obtains a display matrix for which the multidimensional matrix is dimensionally compressed; and a coordinate system association display unit that associates with a coordinate system each component of interest image and each language of the language information of the evaluation object corresponding to the component of interest-position-information based on the display matrix obtained with the display matrix calculation unit, and displays this on the screen.
8. The market evaluation investigation device according to claim 7, wherein in the display matrix calculation unit, the component of interest-position-information and the language information respectively fetched for a plurality of the evaluation objects are consolidated into one multidimensional matrix, and the display matrix is found from the multidimensional matrix.
9. The market evaluation investigation device according to claim 7, wherein in the display matrix calculation unit, an overall image of the evaluation object is added to the same row or column item as the component of interest-position-information in the multidimensional matrix, and a total value of each frequency of the language information used as an element of the multidimensional matrix is used as an element of the overall image, while the display matrix is obtained by dimensionally compressing the multidimensional matrix obtained using that, and in the coordinate system association display unit, in addition to the component of interest image and the language, the overall image is also associated with the coordinate system, and is displayed on the screen.
10. The market evaluation investigation device according to claim 9, wherein the coordinate system association display unit includes a display selection unit of the component of interest image that selectively hides the display of the component of interest image on the screen.
11. The market evaluation investigation device according to claim 7, further comprising an element correlation display unit which, by correlating each item of the component of interest-position-information in the multidimensional matrix with each language of the language information used as an element of each of the items and displaying this on the screen, displays the correlation of the component of interest image and each language displayed on the screen with the coordinate system association display unit.
12. A market evaluation investigation method comprising:
- an evaluation object display step of a computer displaying an image of an evaluation object on a screen;
- a component of interest-position-information fetching step of the computer displaying a component index that indicates in parts the evaluation object on the screen on which the entire image of the evaluation object is displayed, and fetching position information of component of interest by fetching the position information on the screen of the component index moved on the screen by outside input;
- a storage step of the computer storing in an information storage unit the component of interest-position-information fetched with the component of interest-position-information fetching step;
- a grouping calculation step of the computer using the component of interest-position-information stored in the information storage unit as the subject, and executing a calculation process that gathers and groups items with nearby positions of the components of interest; and
- a display step of the computer displaying on the screen the positions of the components of interest made into a small number by being grouped with the grouping calculation step, and displaying a frequency belonging to each of the components of interest by differentiating the display format of the position of the component of interest according to the number of the component of interest-position-information gathered in each component of interest group.
13. The market evaluation investigation method according to claim 12, further comprising:
- a language information fetching step of the computer fetching from outside language information constituted from language representing a reason for interest regarding the component of interest;
- a storage step of the computer affiliating the language information fetched with the language information fetching step to the component of interest-position-information fetched with the component of interest-position-information fetching step and storing this in the information storage unit;
- a display matrix calculation step of the computer using the component of interest-position-information and the language information stored in the information storage unit, finding a multidimensional matrix using as an element each value of a frequency distribution represented having the component of interest-position-information be either of row items or column items and having the language information be the other of the row items or the column items, and finding a display matrix for which the multidimensional matrix is dimensionally compressed; and
- a coordinate system association display step of the computer, based on the display matrix obtained with the display matrix calculation step, associating each component of interest image of the evaluation object corresponding to the component of interest-position-information and each language of the language information with a coordinate system, and displaying that on the screen.
14. An information recording medium readable by a computer on which a program is placed for executing on the computer the market evaluation investigation method comprising: an evaluation object display step of a computer displaying an image of an evaluation object on a screen; a component of interest-position-information fetching step of the computer displaying a component index that indicates in parts the evaluation object on the screen on which an entire image of the evaluation object is displayed, and fetching position information of component of interest by fetching the position information on the screen of the component index moved on the screen by outside input; a storage step of the computer storing in an information storage unit the component of interest-position-information fetched with the component of interest-position-information fetching step; a grouping calculation step of the computer using the component of interest-position-information stored in the information storage unit as the subject, and executing a calculation process that gathers and groups items with nearby positions of the components of interest; and a display step of the computer displaying on the screen the positions of the components of interest made into a small number by being grouped with the grouping calculation step, and displaying a frequency belonging to each of the components of interest by differentiating the display format of the position of the component of interest according to the number of the component of interest-position-information gathered in each component of interest group.
15. A program for executing on a computer the market evaluation investigation method comprising: an evaluation object display step of a computer displaying an image of an evaluation object on a screen; a component of interest-position-information fetching step of the computer displaying a component index that indicates in parts the evaluation object on the screen on which an entire image of the evaluation object is displayed, and fetching position information of component of interest by fetching the position information on the screen of the component index moved on the screen by outside input; a storage step of the computer storing in an information. storage unit the component of interest-position-information fetched with the component of interest-position-information fetching step; a grouping calculation step of the computer using the component of interest-position-information stored in the information storage unit as the subject, and executing a calculation process that gathers and groups items with nearby positions of the components of interest; and a display step of the computer displaying on the screen the positions of the components of interest made into a small number by being grouped with the grouping calculation step, and displaying a frequency belonging to each of the components of interest by differentiating the display format of the position of the component of interest according to the number of the component of interest-position-information gathered in each component of interest group.
Type: Application
Filed: Apr 3, 2012
Publication Date: Apr 3, 2014
Applicant: CREATIVE BRAINS, K.K. (Tokyo)
Inventor: Kazuhiko Suzuki (Tokyo)
Application Number: 14/119,118
International Classification: G06Q 30/02 (20060101);